Just to provide some BASIC items of differences on data sources for home/work FLOWS.
ACS and CTPP based on ACS: This is a sample SURVEY of households and household members.
The ACS is an address-based sample. The response rate to ACS is lower than to Census 2000
"long form" and the field follow-up for non-respondents to the mail-back form is
for a SAMPLE of non-respondents. All the responses are weighted to come up with a total
estimate. Respondents are asked to provide their workplace location, which are then
geocoded by the Census Bureau. About 25 % of workers do not provide sufficient
information to code to a census block and their workplace is imputed using a routine using
4 variables: industry, occupation, travel mode and travel speed. See the recent CTPP
Status Report article by Melanie Rapino on the details of the imputation process.
LEHD: is from ADMINISTRATIVE records, based on Employment Security (unemployment), and
augmented with federal employment records. The base is from files submitted to the Census
Bureau from State Departments of Employment Security who maintain the QCEW files
(formerly called ES-202). LEHD does not currently include self-employed, or military
employment. In this case, the workplace location is the address(es) provided by
businesses. If the business has multiple locations in a county, many states require that
each location be reported. Only in Minnesota is the employee record tied to a specific
workplace location. That is, in all other states, for businesses with multiple sites, the
link between an employee's home location and the workplace location is imputed. That
is, an algorithm selects among the multiple worksites which is the most likely workplace
for that worker. The home location may be moved to protect individual confidentiality,
but largely the moves are within the same census tract. The link between the unemployment
(QCEW) and home location is primarily from IRS tax records, using an encrypted version of
SSN.
Sometimes people use a different home address for IRS than where they are actually living,
for example young people or college students may use their parents' address. And
similarly, sometime people report on the ACS form a workplace location where they may not
actually be working, for example, a construction worker might report the business address
of his employer. So, there is plenty of room for inconsistencies.
Private lists of businesses and employment totals: Typically these are compiled using web
resources and telephone surveys. Often the list guarantees that each business has been
contacted within 3 (or 5) years to get the most current employment figures. Because
these firms have been compiling these lists and checking them for so many years, they have
been able to refine their databases. However, these lists do not provide FLOW between
work and home, just workplace location totals.
Please let me (and the listserv) know if I have made any mistakes above.
Elaine Murakami
________________________________________
From: ctpp-news-bounces(a)chrispy.net [ctpp-news-bounces(a)chrispy.net] on behalf of
hprawiranata
mitcrpc.org [hprawiranata(a)mitcrpc.org]
Sent: Friday, March 08, 2013 11:49 AM
To: ctpp-news(a)chrispy.net
Subject: Re: [CTPP] Commuter Flow Data Concerns
Thanks, Ed.
I have been working LEHD data for 2 weeks and comparing it with other
sources (REMI, Claritas, WP, etc). All sources has different number
for 2010, for example, which one is the good one ? LEHD has all jobs,
primary jobs, private jobs, etc. I found the use primary job data give
close number with other sources, am I on the right track ?
I have not compared LEHD data with 'just released' commuting data set,
I guess it would be the same and LEHD has better resolution. What do
you think ?
Thanks,
Hary
On Fri, Mar 8, 2013 at 10:22 AM, Ed Christopher <edc(a)berwyned.com> wrote:
A few of us have been getting calls and emails from
people who have been
looking at the commuter follow data the Census Bureau just released, the
On the Map (LEHD) and some of the earlier flow data. If you have, you
probably noticed some interesting (or maybe wierd) things. Needless to
say a good understanding of the data is needed to sort out some of the
issues. Below are some references and points that may be of interest.
What are people noticing with the data?
---------------
In the January 20011 Status Report newsletter there is an article from
the NY State DOT by the late Nathan Erlbaum which discusses the issues
surrounding the various flow data sets. The article “Commutation Flow:
CTPP 2000, ACS and CTPP, and LEHD-OTM” can be found at
http://www.trbcensus.com/newsltr/sr0111.pdf (starting on page 6). In it
is a link to a larger body of work that Nathan did on the topic. a read
of this is must if you are having questions about the data.
Regarding the LEHD and CTPP flow data, there is an NCHRP report that has
been completed that is also worth reviewing. It gets into detail about
some of the differences between the two data sets. It can be found at
http://onlinepubs.trb.org/onlinepubs/nchrp/docs/NCHRP08-36%2898%29_FR.pdf
Elaine Murakami reminds me that
1. LEHD does not include self-employed (approx. 10%)
2. LEHD only recently included federal workers, but does not include
MILITARY
3. LEHD is jobs and ACS is "workers at work".
Periodically issue like this pop up and there is a discussion on the
list serve. there is an archive but off course it takes some time to
mine and search through. If anyone is interested in looking through it,
it is open to the public and available at
http://www.chrispy.net/pipermail/ctpp-news/. There was a discussion
around august of 2010 dealing with flow data but it centered more on
availability.
Hope this helps.
--
Ed Christopher
708-283-3534 (V)
708-574-8131 (cell)
FHWA RC-TST-PLN
4749 Lincoln Mall Drive, Suite 600
Matteson, IL 60443
_______________________________________________
ctpp-news mailing list
ctpp-news(a)ryoko.chrispy.net
http://ryoko.chrispy.net/mailman/listinfo/ctpp-news
_______________________________________________
ctpp-news mailing list
ctpp-news(a)ryoko.chrispy.net
http://ryoko.chrispy.net/mailman/listinfo/ctpp-news